A method for efficient autonomous driving planning includes receiving a current driving-scene data and a predicted driving-scene data. The current driving-scene data is indicative of a current driving scene around a host vehicle. The predicted driving-scene data is indicative of a predicted driving scene around the host vehicle. The predicted driving scene around the host vehicle is different from the current driving scene around the host vehicle. The method further includes converting the current driving-scene data and the predicted driving-scene data into a first scene-graph and a second scene-graph, respectively. The method further includes determining a plurality of scene change metrics using the first scene-graph and the second scene-graph. The method further includes selecting between a first trajectory planning process and a second trajectory planning process based on the plurality of scene change metrics.
Legal claims defining the scope of protection, as filed with the USPTO.
2. The method of claim 1, wherein the plurality of scene change metrics includes a graph edit distance between the first scene-graph and the second scene-graph, determining the plurality of scene change metrics using the current driving-scene data and the predicted driving-scene data includes determining the graph edit distance between the first scene-graph and the second scene-graph, and selecting between the first trajectory planning process and the second trajectory planning process based on the plurality of scene change metrics includes selecting between the first trajectory planning process and the second trajectory planning process based on the graph edit distance between the first scene-graph and the second scene-graph.
4. The method of claim 3, wherein the plurality of scene change metrics includes a Mahalanobis distance between a predicted position of a remote actor and a current position of the remote actor, the predicted position of the remote actor is part of the predicted driving-scene data, the current position of the remote actor is part of the current driving-scene data, determining the plurality of scene change metrics using the current driving-scene data and the predicted driving-scene data includes determining the Mahalanobis distance between the predicted position of the remote actor and the current position of the remote actor, and selecting between the first trajectory planning process and the second trajectory planning process based on the plurality of scene change metrics includes selecting between the first trajectory planning process and the second trajectory planning process based on the Mahalanobis distance between the predicted position of the remote actor and the current position of the remote actor.
6. The method of claim 5, wherein the plurality of scene change metrics includes a target lane shape, determining the plurality of scene change metrics using the current driving-scene data and the predicted driving-scene data includes determining the target lane shape, and selecting between the first trajectory planning process and the second trajectory planning process based on the plurality of scene change metrics includes selecting between the first trajectory planning process and the second trajectory planning process based on the target lane shape.
8. The method of claim 7, further comprising determining a sum of weighted scene change metrics.
10. The method of claim 9, further comprising determining whether the sum of the weighted scene metrics is greater than a predetermined threshold.
11. The method of claim 10, wherein selecting between the first trajectory planning process and the second trajectory planning process based on the plurality of scene change metrics includes selecting the first trajectory planning process in response to determining that the sum of the weighted scene metrics is greater than the predetermined threshold.
12. The method of claim 10, wherein selecting between the first trajectory planning process and the second trajectory planning process based on the plurality of scene change metrics includes selecting the second trajectory planning process in response to determining that the sum of the weighted scene metrics is not greater than the predetermined threshold.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
September 20, 2022
October 29, 2024
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.